Dynamic downscaling of global climate projections for Eastern Europe with a horizontal resolution of 7 km
- First Online:
- Cite this article as:
- Pavlik, D., Söhl, D., Pluntke, T. et al. Environ Earth Sci (2012) 65: 1475. doi:10.1007/s12665-011-1081-1
- 264 Downloads
Climate change is one of the key factors influencing the quantity and quality of water resources in hydrologically sensitive regions. In order to downscale global climate simulations from horizontal resolutions of about 125–200 km to about 7 km, a double nesting strategy was chosen. The modelling approach was implemented with the Regional Climate Model CCLM (COSMO-Climate Local Model) with a first nesting covering a central part of Europe and with a second nesting covering parts of Poland, Belarus, and the Ukraine. A control run—driven by global reanalysis data—was evaluated by comparing the model results with corresponding reference data. Long-term yearly and monthly mean differences of temperature and precipitation were statistically assessed. As reference data for the first nesting, the gridded CRU data set with a horizontal resolution of about 55 km was used. Station data of the NOAA and ECA databases were interpolated to provide an appropriate reference data set for the second nesting. Both nestings overestimated long-term yearly precipitation means. Seasonal evaluation of the first nesting showed positive precipitation biases for spring and winter months and negative biases in summer. Furthermore, differences in the spatial precipitation patterns occured, especially in the high mountain area of the Carpathians. The second nesting overestimated precipitation in spring and summer with smaller biases than in the first nesting. Long-term area means of temperature were properly reproduced. The first nesting showed an overestimation for all months with maximal deviations in summer and spring. In contrast, the second nesting was slightly too warm for summer and autumn and too cold for winter and spring.